Proc. Natl. Acad. Sci. USAVol 74, No. 4, pp. 1711-1715, April 1977Medical Sciences

Radioimmunoassay of creatine kinase-B isoenzyme in human sera:Results in patients with acute myocardial infarction

(myocardial necrosis/chest pain/coronary artery disease)

JAMES T. WILLERSON*, MARVIN J. STONE, RUBY TING, AMAL MUKHERJEE, CELSO E. GOMEZ-SANCHEZ,PATRICIA LEWIS, AND LOUIS B. HERSHDepartments of Internal Medicine (Cardiac Unit and Evelyn L. Overton Hematology-Oncology Laboratory), and Biochemistry, University of Texas HealthScience Center at Dallas, Dallas, Texas 75235

Communicated by Eugene Braunwald, December 27,1976

ABSTRACT A radioimmunoassay was developed to mea-sure serum levels of the B isoenzyme of creatine kinase (ATP:creatine N-phosphotransferase, EC 2.7.3.2) (CPK) in order toevaluate the time course and frequency of MB isoenzyme ele-vation in patients with acute myocardial infarction. The methodcan identify as little as 0.2 ng of the B portion of the CPK-MBisoenzyme, does not significantly crossreact with CPK-MMisoenzyme, and is not affected by storage of serum at -20°. CPKisoenzyme containing B subunits was detected in 48 out of 51sera from normal adults; serum levels in these individualsranged between 1.2 and 12.5 ng/ml [mean + SEM was 2.7 i 0.30ng/mlJ. The mean serum level of CPK-B isoenzyme in a pool ofsera obtained from 100 normal subjects was 2.9 ± 0.35 ng/ml;two patients with rhabdomyolysis that were studied had serumCPK-B isoenzyme levels of 2.5 and 3.5 ng/ml, respectively. Incontrast, serum levels of the CPK-B isoenzyme were markedlyelevated in sera from 18 patients with acute myocardial infarctswhen obtained within 12 hr after hospital admission; the meani SEM concentration was 56 + 7.8 ng/ml. We performed serialdeterminations on 14 patients with acute myocardial infarctsand demonstrated that maximal serum CPK-B levels occurredwithin the first 12 hr after admission and were lower thereafter.The serum concentration of B-containing CPK isoenzyme in 19additional patients admitted with chest pain but without acutemyocardial infarction was 3.4 ± 0.50 ng/ml. Thus, radioim-munoassay measurement of CPK-B isoenzyme appears to be auseful and sensitive test for the detection of acute myocardialinfarcts in patients.

The diagnosis of acute myocardial infarction is ordinarily madeon the basis of a characteristic clinical history, and typical serumenzyme and electrocardiographic alterations. Recently, myo-cardial scintigraphy has also been utilized to aid in the detectionof both acute transmural and nontransmural myocardial in-farcts (1). In some patients, however, these data are insufficientto provide an exact diagnosis of acute myocardial infarctioneither because of temporal considerations or relative non-specificity of the observations (1, 2). More recently, it wassuggested that measurement of the "myocardial-specific"creatine kinase (ATP:creatine N-phosphotransferase, EC2.7.3.2) (CPK) isoenzyme, the MB isoenzyme, helps to distin-guish between patients with acute myocardial infarcts and thosewith chest pain of other etiology (3, 4). CPK exists as three iso-enzymes and these are BB which is present in brain, MB whichprimarily exists in heart muscle, and MM which is present bothin heart and skeletal muscle (3, 4). It has been previously shownthat intramuscular injections and other forms of skeletal muscletrauma may elevate total serum CPK because of MM isoenzymerelease from skeletal muscle, thus making it difficult to relysolely on measurements of total serum CPK to establish a di-

agnosis of acute myocardial infarction in patients. Methodspreviously available for measuring the myocardial specificCPK-MB isoenzyme included electrophoretic and columnseparation techniques (3-5), but we have experienced somedifficulty with reproducibility of measurements and precisequantitation of the CPK-MB isoenzyme at the lower levels ofenzyme concentrations when we utilized these methods. Spe-cifically, the electrophoretic method (4) that was used is onlyqualitative and the column separation technique (5) itself, witheither spectrophotometric or fluorometric determinations,generally lacks the necessary sensitivity and consistency at thelower levels of CPK-MB measurement that would be expectedof a sensitive radioimmunoassay. It should also be noted thatthe CPK-MB isoenzyme accounts for only approximately10-15% of the total CPK present in heart muscle, so precisequantitation for accuracy of measurements is of primary im-portance in measuring this CPK isoenzyme. For these reasons,we have developed a double antibody competitive-bindingradioimmunoassay which detects the CPK-B isoenzyme todetermine if a sensitive, specific, and quantitative assay for theB portion of the CPK myocardial-specific isoenzyme would beclinically useful.

METHODSPreparation of BB Isoenzyme of CPK. Brain obtained from

humans at post-mortem examination was utilized to separateand purify the brain specific CPK isoenzyme (BB isoenzyme).The technique utilized was a modification of the one describedby Keutel et al. (6). The purified BB isoenzyme was lyophilizedand stored in small aliquots at -20°.

Several other proteins were utilized for control purposes (seeResults). The MM and MB isoenzymes of CPK were preparedfrom human skeletal muscle and heart, respectively, by themethod of Carlson et al. (7). Citrate (si)-synthase [citrate ox-aloacetate-lyase (CoA-acetylating), EC 4.1.3.7] from humanheart was prepared by the method of Mukherjee and Srere (8).Human heart myoglobin was prepared by utilizing methodswe have previously described (2). CPK-MM isoenzyme fromrabbit skeletal muscle was obtained from Sigma Chemical Co.,St. Louis, MO. Phosphorylase b, glycogen phosphorylase (a-1,4-glucan:orthophosphate a-glucosyl transferase, EC 2.4.1.1)from human heart was prepared by a modification of theDavies et al. method (9).Determination of Enzyme Activity. Creatine kinase activity

was determined spectrophotometrically by using the coupledenzyme assay as described by Rosalki (10). Calbiochem (SanDiego, CA) CPK reagents were used. One unit of enzyme wasdefined as 1 ,umol of creatine phosphate consumed per min at25°0

Criteria of Homogeneity of CPK-BB Isoenzyme. Polyac-1711

Abbreviation: CPK, creatine kinase.* To whom reprint requests should be addressed at: Ischemic HeartCenter, L5-134, The University of Texas Health Science Center atDallas, 5323 Harry Hines Boulevard, Dallas, Texas 75235.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 26

, 202

0

1712 Medical Sciences: Willerson et al.

rylamide disc gel electrophoresis of the enzyme was carried outaccording to the method of Davies (11). Electrophoresis wasperformed in 7.5% polyacrylamide gel in 0.05 M Tris-glycinebuffer, pH 8.3, containing 15 mM 2-mercaptoethanol. For so-dium dodecyl sulfate gel electrophoresis of the enzyme, theprocedure of Weber and Osborn (12) was used with 10%polyacrylamide gels and a running buffer of 0.05 M sodiumphosphate/0.1% sodium dodecyl sulfate, pH 7.0. The gels werestained in 0.2% Coomassie brilliant blue.

Protein was assayed in column eluates by using the methodof Warburg and Christian (13). The protein of pure prepara-tions was determined by the method of Lowry et al. (14) withcrystalline bovine-serum albumin as a standard.

Immunization of Rabbits. Antisera against theBB isoen-zyme of CPK were prepared by immunizing adult albinorabbits. The antigen (0.5 mg) was emulsified in completeFreund's adjuvant and injected in multiple subcutaneous sitesat 3-4 weekly intervals. The rabbits were bled 10-14 days aftereach booster injection and the serum was stored at-20° untilused. Periodic testing of the rabbit sera with Ouchterlonydouble-diffusion analysis continued until a clearly recognizablesingle band of antibody versus CPK-BB antigen was identi-fied.

Radiolabeling of CPK-BB Isoenzyme. CPK-BB isoenzymewas iodinated by conjugate labeling as described by Bolton andHunter (15). LyophilizedBB isoenzyme (2.3 mg) was dissolvedin 0.4 ml of 0.1 M sodium borate buffer (pH 8.5) to yield aconcentration of 5 mg of BB isoenzyme per ml of borate buffer.The CPK was then dialyzed in the same buffer before radio-labeling.

For the radiolabeling, 1 mg of N-hydroxysuccinimidyl-3-(4-hydroxyphenyl) propionate (Tagit) (Calbiochem) was dis-solved in 50 ml of benzene; a tube containing 10 ,l of this so-lution was taken to dryness under reduced pressure. Ten mi-croliters of 125I [specific activity 100 mCi/ml (Amersham/Searle Corp., Arlington Heights, IL)], 10 Al of ester (Tagit) (1mg/50 ml of benzene), and 10,ul of chloramine-T (50 Mg)(Eastman Organic Chemical Co., Rochester, NY) were added.The reaction was terminated by the addition of 120 Mg (10,ul)of sodium metabisulfite with KI as a carrier and N,N-dimeth-ylformamide. The labeled product was extracted into 0.5 mlof benzene twice and slowly air-dried. Three micrograms ofthe BB isoenzyme of CPK (15 ,l of 0.2 Mig/ml stock) was addedand incubated in an ice-bath for 2 hr. Next, 0.1 ml of 0.2 Mglycine was added and the mixture put over a 10-ml SephadexG-50 column. The final specific activity for the iodinated BBisoenzyme of CPK was 2.6 X 106 cpm/,ug (2.1 X 105) cpm/pmol; this calculated specific activity assumes total recoveryof the CPK-BB isoenzyme from the column and homogeneouslabeling of the protein.Radioimmunoassay Procedure. All determinations (stan-

dards and unknowns) were performed in 12 X 75 mm silicon-treated tubes containing 0.05 M borate buffer, pH 7.8, in thepresence of 0.1% sodium azide and 5% normal human serum(prepared by clotting outdated plasma obtained from the bloodbank at Parkland Memorial Hospital, Dallas, TX). The additionof a constant amount of 125I-BB CPK isoenzyme to serial dilu-tions of antiserum obtained from bleeding a single rabbit (no.17) disclosed maximum binding of isoenzyme to antibody atan antiserum dilution of 1:4500; this dilution was utilized in allsubsequent assays. Approximately 6000 cpm of 125I-BB CPKisoenzyme were present in each tube. A 20-Ml aliquot of eachserum sample (from normal subjects or patients) was routinelyassayed. For those samples with a CPK-B value on the flatportion of the standard curve, the determination was repeated

A B

1-...

FIG. 1. (A) Disc gel electrophoresis of CPK-BB isoenzyme pu-rified from brain. Twenty micrograms of enzyme were loaded on thegel. There is a single band near the bottom of the gel that representsthe purified CPK-BB isoenzyme. (B) Sodium dodecyl sulfate/gelelectrophoresis of CPK-BB isoenzyme from brain. Fifteen microgramsof enzyme were loaded on this gel. The single band near the middleof the gel represents the purified CPK-BB isoenzyme.

with 10 ,l of serum. Total volume per tube was 0.6 ml. All de-terminations were performed in triplicate and a standard curvewas performed each day of testing. In these experiments, aninitial 24-hr period of incubation at 40 of unlabeled BB isoen-zyme with the antiserum optimized the immunoassay resultsand therefore this period of incubation was routinely employedbefore the addition of 125I-labeled CPK-BB isoenzyme. Sepa-ration of unreacted enzyme from antibody-bound CPK-BBisoenzyme was accomplished by the double-antibody methodutilizing goat antiserum against rabbit IgG (Miles Laboratory,Elkhart, IN). The second antibody (0.3 mg per tube 0.1 ml ofassay volume) was added 24 hr after the addition of labeledCPK and the precipitates were assayed for radioactivity afterincubation at 40 for another 24 hr. Thus, the radioimmunoassayutilized in these studies took 3 days to complete. More recently,however, we have found that the 24 hr of initial incubation isnot absolutely necessary for optimal use of the radioimmu-noassay and that the time required to complete the test canalready be shortened to approximately 36 hr. Further reduc-tions in the time required to complete the immunoassay can beexpected with the use of higher-affinity antibodies and by usingammonium sulfate to separate bound from free CPK-B en-zyme.

Using a specific activity of CPK-MB isoenzyme of 244 (6),we find this radioimmunoassay is approximately 80 to 85 timesmore sensitive than the spectrophotometric measurements and

Proc. Natl. Acad. Sci. USA 74 (1977)

Dow

nloa

ded

by g

uest

on

Nov

embe

r 26

, 202

0

Proc. Natl. Acad. Sci. USA 74 (1977) 1713

c

0-oC0

0-

60O

55

_

50 _,

45_

40_

60

50

40 >

20

10 _

0.2 0.5 1.0 2.5 5 10 25 50 NSUnlabeled CPK-BB Isoenzyme, ng

FIG. 2. A typical standard curve for the CPK-BB radioimmu-noassay. As little as 0.2 ng of CPK-B isoenzyme can be recognized byusing this immunoassay. The small crossbars demonstrate the rangesof triplicate determinations. NS refers to normal subjects.

5 to 10 times more sensitive than a standard fluorometricassay.Serum Samples Analyzed. The presence of CPK-B isoen-

zyme units in sera from 51 normal adults was determined.Determinations on pooled sera obtained from 100 normalsubjects (medical students) and stored at -20° for 3 years weremade. Fresh serum samples from seven normal controls werealso tested. Sera from patients with chest pain of varying eti-ology admitted to the Coronary Care Unit at Parkland Me-morial Hospital, Dallas, TX, between 6 and 12 months ago, alsowere analyzed. Serial serum samples were obtained from 16tof these patients for analysis. Serum samples were also analyzedfrom 19 patients with chest pain that subsequently turned outnot to be due to acute myocardial infarction. The etiology ofthe chest pain in most of these 19 patients was angina pectoris.Sera were also available from two patients with rhabdomyolysisbut without acute myocardial infarcts. The patients evaluatedin the present studies neither had clinical evidence of cere-brovascular injury or any other form of brain disease, nor hadany of these patients undergone electrical defibrillation beforeobtaining their blood samples. Verbal informed consent wasobtained for the venous blood collections. All patients' sera werestored at -20° before being tested.

Identification of Patients with Acute Myocardial Infarcts.The recognition of acute myocardial infarcts in patients wasaccomplished when all of the following occurred: (i) a historyof severe and prolonged typical substernal chest pain; (ii)classical changes in the QRS waves that evolved in the expectedmanner for transmural infarcts or ST-T wave changes consistentwith a diagnosis of acute subendocardial infarction; (iii) typicalelevation and subsequent evolution of the total serum CPK inthe absence of any form of skeletal muscle damage or intra-muscular injections; and (iv) a positive myocardial scintigramwith technetium-99m stannous pyrophosphate that developedwithin 48 hr after the onset of severe chest pain.

RESULTSThe final specific enzyme activity of the purified CPK-BBisoenzyme varied from 260-280 units/mg protein at 25°. Totalrecovery of the CPK-BB isoenzyme was 40-45% of the initialactivity. Fig. 1A shows a single band of protein on disc gelelectrophoresis. Sodium dodecyl sulfate/gel electrophoresis ofthe enzyme also shows a single protein band confirming thehomogeneity of the protein (Fig. 1B).

t Fourteen of these patients had acute myocardial infarcts and two hadunstable angina pectoris.

0,m

cL

to;

0~

U

E0Cl,

35 _

30 _

25 _

20 _

15_

10

5

I

INormals Chest pain Acute(n = 44) but no myocardial

infarct infarcts(n= 19) (n= 18)

FIG. 3. The serum CPK-B isoenzyme levels as determined byradioimmunoassay in normal individuals, those with chest pain butwithout acute myocardial infarction, and those with acute myocardialinfarcts. The bars show mean serum values and the crossbars showthe standard errors. The standard deviations (4ng/ml) for each groupare as follows: 2.3 for normals, 2.2 for patients with chest pain but noinfarcts, and 33.0 for the patients with acute myocardial infarcts.

A typical standard curve, derived by the addition of knownamounts of nonradioactive CPK-BB isoenzyme to tubes con-taining constant amounts of antibody and labeled antigen, isshown in Fig. 2. Using this standard curve, we can easily detecta concentration of 0.2 ng of enzyme per tube; six such curvesrun under similar conditions were essentially identical.

Precision, Reproducibility, and Specificity. Within a singleassay, values with triplicate determinations agreed within 5%.Repeated freezing and thawing over 72 hr had no effect onthese assay values. The addition of known amounts (1, 2.5, 5,10, and 20 ng) of unlabeled CPK-BB isoenzyme to normalserum in four separate experiments resulted in average recoveryof 94 + 3.2% (SEM) and indicated that no systematic error couldbe detected. Crossreactivity to other myocardial constituentsrevealed that only 0.01% of human heart phosphorylase b,0.032% of human heart citrate (si)-synthase, 0.004% of humanheart myoglobin, and 0.0009% of human skeletal muscleCPK-MM isoenzyme were detectable with this assay. Nomeasurable crossreactivity with CPK-MM isoenzyme fromrabbit skeletal muscle was identified. The lack of significantreactivity of CPK-MM isoenzyme from human skeletal musclepoints to the specific recognition of CPK-MB or BB isoenzymesin this immunoassay. Four separate experiments, performedto determine the cross-reactivity of the antibody to CPK-BBisoenzyme for human-heart CPK-MB isoenzyme revealed thatan average of 44% (range 35-59%) of the serum level for thisCPK isoenzyme was detected. Some variation in the recognitionof CPK-MB by our antibody should be expected when otherCPK-MB isoenzyme preparations are tested and different

Medical Sciences: Willerson et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 26

, 202

0

1714 Medical Sciences: Willerson et al.

.-' F

A7 1

86

-5

V3 ~ ~ \

~~~~~1~~.CI)

0 4 8 12 16 20 24 28 32 36 40 44 48 54 58 72 90 94 98

Time after hospital admission, hr

1,

-I

- <1*

LLI

I

LA

I-I

.I..I

3- i 2 4 6 58 20 22 24 26 23

T' ~l t' rim Or.).-%.I.''; Ki.

C

C

so ---/-

0 48121620162 42 8328 3 64064044834 56789167 90 1415

Hours after hospital admission Days after admission

FIG. 4. (A) Serial CPK-B isoenzyme levels in two patients with unstable angina pectoris, but without acute myocardial infarcts. Note that

the serum levels in both of these patients did not exceed a value of 6 ng/ml. (B) The mean values and standard errors for serum CPK-B isoenzyme

levels in the 14 patients with acute myocardial infarcts. Some of the time periods have less than 14 patients. (C) Serial serum CPK-B isoenzyme

levels in 14 patients with acute myocardial infarcts are shown.

antibody fractions are evaluated. The lack of significantcross-reactivity with any of the other myocardial cellular con-stituents tested, and the recognition of approximately half theadded amount of CPK-MB isoenzyme from human heart fur-ther confirms that the antibody used in this assay is specific forthe B portion of the CPK-MB isoenzyme.Serum B Levels in Normal Persons and Patients. Sera ob-

tained from normal adults contained less than 12.6 ng/ml ofCPK-B isoenzyme. The mean value for the 44 normal adultswas 2.9 0.3 ng/ml with a range of from less than 1 to 12.5ng/ml (Fig. 3). The mean serum level of CPK-B isoenzymeobtained in an additional seven normal individuals and assayedwhile "fresh" was 1.8 0.07 ng/ml. In pooled sera obtainedfrom 100 normal subjects, the CPK-B isoenzyme level was 2.9± 0.35 ng/ml. A value in excess of 12.5 ng/ml was taken as

abnormal because we have not yet encountered higher serumlevels in normal individuals.

Fig. 3 shows that the serum CPK-B isoenzyme levels in pa-

tients admitted to the hospital with chest pain and from whomserum samples were obtained. Patients with acute myocardialinfarcts had abnormal levels of serum CPK-B isoenzyme, butthose with chest pain of other etiology had normal values (Fig.4A). However, it should be noted that elevations of serum

CPK-B isoenzyme persist in some patients for longer periodsof time (Fig. 4B). Fig. 4C reveals that serum CPK-B isoenzymelevels rise in patients with acute myocardial infarcts within 4-5hr after admission to the hospitalt and return toward normalvalues ordinarily within 24 hr after infarction.

It is interesting to compare the time for increase of the serumCPK-B isoenzyme levels detected by our radioimmunoassaywith results we have previously obtained by utilizing a ra-

dioimmunoassay for serum myoglobin in patients with in-farction (2). Fig. 4 shows that serum CPK-B isoenzyme levelsincrease at approximately similar times (but in some individualpatients slightly later) and reach a peak value later than was

found for serum myoglobin values in patients with acutemyocardial infarcts.

DISCUSSIONThe data obtained in this study document the development ofa specific and sensitive radioimmunoassay for CPK-B isoen-zymes; this method allows detection of as little as 0.2 ng of theB containing CPK isoenzyme in sera obtained from patients.

$ In almost every instance, patients came to the hospital within 3 hrof the onset of chest pain.

L i

L11R 30 3 ? 34

_0.4 0AYS

Proc. Natl. Acad. Sci. USA 74 (1977)

-.,

Dow

nloa

ded

by g

uest

on

Nov

embe

r 26

, 202

0

Proc. Natl. Acad. Sci. USA 74 (1977) 1715

The measurements made in sera obtained from 51 normal in-dividuals and pooled sera obtained from 100 normal subjectssuggest that serum values above 12.5 ng/ml are abnormal be-cause values above this level for normal controls were not foundin either fresh sera, stored frozen sera, or frozen and thawedsera. There was also no evidence that our antibody to CPK-Bisoenzyme significantly crossreacts with citrate (si)-synthasefrom human heart, heart phosphorylase b, myoglobin, orCPK-MM isoenzyme. Others have also recently reported thedevelopment of a radioimmunoassay for the CPK-MB isoen-zyme but detailed clinical results have not yet been described(16).The results obtained in the present study also demonstrate

and confirm that patients with acute myocardial infarction havemarked elevations of CPK-B isoenzyme (3, 4) levels beginningapproximately 2-6 hr after admission to the hospital. Othershave previously noted the strong association between CPK-MBisoenzyme elevations and the presence of acute myocardialinfarction by using less sensitive and less quantitative mea-surements (3, 4). The elevated B-portion levels of CPK isoen-zyme noted in our patients with acute myocardial infarcts de-creased toward normal levels within 24 hr after admission inthe majority of patients studied but a few individuals retainedelevated serum levels for longer periods of time. Thus, it isimportant to obtain serum from patients within a few hoursafter their hospital admission and to obtain serial serum samplesfor CPK-B analyses in any patient suspected of having an acutemyocardial infarction, if this assay is to be utilized to its fulladvantage. Patients with chest pain not due to myocardial in-farction appear to have normal serum CPK-B isoenzyme levelsby this radioimmunoassay.

It has previously been suggested that measurements of serumcreatine kinase activity, and specifically the "myocardial-specific" CPK isoenzyme (MB isoenzyme), are useful inquantitating the extent of acute myocardial infarcts in exper-imental animals and man (17, 18). The quantitative assay de-veloped in this study should prove helpful in making preciseand sensitive measurements of the B portion of the CPK"myocardial specific" isoenzyme, and thus prove useful infurther refining the quantitative enzymatic assessment of theextent of myocardial injury. Because the antibody utilized inthis radioimmunoassay measures 44% of the CPK-MB isoen-zyme present in patients's serum samples, the apparent amountof CPK-MB isoenzyme present will have to be multiplied bya factor of 2 to determine the exact concentration of CPK-MBisoenzyme present.We have previously shown that a radioimmunoassay mea-

surement of serum myoglobin levels is also of considerable helpin the recognition of acute myocardial infarcts in patients (2).Our assay for serum myoglobin, however, is somewhat limitedby the fact that cardiac and skeletal muscle myoglobins areimmunochemically identical and a number of conditions areassociated with liberation of myoglobin from skeletal muscle(2, 19); thus, in some patients with marked renal insufficiencyand those with extensive skeletal muscle damage either fromtrauma, shock, or neuromyopathic processes, it will be difficultto determine the presence or absence of acute myocardial in-farction by utilizing the radioimmunoassay for serum myo-globin. However, the present immunoassay for B-containingCPK isoenzyme should further help distinguish between thepresence and absence of a myocardial infarction.

Because the radioimmunoassay developed in this study assaysfor the CPK-B isoenzyme, it may also prove useful for recog-nizing and possibly quantitating the extent of brain damage inpatients. In preliminary studies performed with our radioim-munoassay for CPK-B isoenzyme, we have already found a fewpatients with intracranial hemorrhages that have developedincreased cerebrospinal fluid and serum CPK-B levels. This isa matter that will have to be carefully studied further in thefuture; should elevated CPK-B isoenzyme levels be found withthe radioimmunoassay determination in most patients withcerebrovascular accidents, primary central nervous systemdisease, brain tumors, etc., then these clinical situations willrepresent limitations in the detection of acute myocardial in-farction in patients who simultaneously have both clinicalproblems.

The authors wish to acknowledge the technical assistance of Ms.Linda Stevenson, Ms. Andrea Wangler and Mrs. Carolyn Michnoff,and the secretarial assistance of Ms. Donna Place and Mrs. BelindaLambert. We also wish to express our appreciation to the medical houseofficers and nurses in the Coronary Care Unit at Parkland MemorialHospital in Dallas, Texas, without whose help these studies could nothave been performed. This work was supported by the Ischemic HeartDisease Specialized Center of Research (SCOR) Grant HL-17669,Grants HL-15522 and AM13443 from the National Institutes of Health,American Heart Association Grant 75-819, and Robert A. Welch Grant1391. Dr. Willerson is an Established Investigator of the AmericanHeart Association. Dr. Hersh is a recipient of a Research Career De-velopment Award no. GM70239.

1. Willerson, J. T., Parkey, R. W., Bonte, F. J., Meyer, S. L. Atkins,J. M. & Stokely, E. M. (1975) Circulation 51, 1046-1052.

2. Stone, M. J., Willerson, J. T., Gomez-Sanchez, C. E. & Waterman,M. (1975) J. Clin. Invest. 56, 1334-1339.

3. Roberts, R. & Sobel, B. E. (1973) Ann. Int. Med. 79,741-743.4. Wagner, G. S., Roe, C. R., Limbird, L. E., Rosati, R. & Wallace,

A. G. (1973) Circulation 47,263-269.5. Henry, P. D., Roberts, R. & Sobel, B. E. (1976) Clin. Chem. 21,

844-849.6. Keutel, H. J., Okabe, K., Jacobs, H. K., Ziter, F., Maland, L. &

Kuby, S. A. (1972) Arch. Biochem. Biophys. 150,648-678.7. Carlson, E., Roberts, R. & Sobel, B. E. (1976) J. Mol. Cell. Cardiol.

8, 159-167.8. Mukherjee, A. & Srere, P. A. (1976) J. Biol. Chem. 251, 1476-

1480.9. Davies, C. H., Schliselfeld, L. H., Wolf, D. P., Leavitt, C. A. &

Krebs, E. G. (1967) J. Biol. Chem. 242, 4824-4833.10. Rosalki, S. B. (1967) J. Lab. Clin. Med. 69,969-705.11. Davies, B. J. (1964) Ann. N.Y. Acad. Sci. 121, 404-427.12. Weber, K. & Osborn, M. (1969) J. Biol. Chem. 244, 4406-

4412.13. Warburg, 0. & Christian, W. (1942) Biochem. Z. 310, 384-

421.14. Lowry, 0. H., Rosebrough, N. J., Farr, A. L. & Randall, R. J.

(1951) J. Biol. Chem. 193,265-275.15. Bolton, A. E. & Hunter, W. M. (1973) Biochem. J. 133, 529-

539.16. Roberts, R., Sobel, B. E. & Parker, C. W. (1976) Science 194,

855-857.17. Shell, W. E., Kjekshus, J. K. & Sobel, B. E. (1971) J. Clin. Invest.

50,2614-2625.18. Sobel, B. E., Bresnahan, G. F., Shell, W. E. & Yoder, R. D. (1972)

Circulation 46, 640-648.19. Kagen, L. J. (1973) in Myoglobin (Columbia University Press,

New York), pp. 79-116.

Medical Sciences: Willerson et al.

Dow

nloa

ded

by g

uest

on

Nov

embe

r 26

, 202

0